REDUCING THE TIME OF TASK EXECUTION WITH EXISTING

Int j simul model 16 (2017) 3, 484-496

ISSN 1726-4529 Original scientific paper

https://doi.org/10.2507/IJSIMM16(3)10.394 484

REDUCING THE TIME OF TASK EXECUTION WITH

EXISTING RESOURCES – COMPARISON OF APPROACHES

Zupancic, D.*; Buchmeister, B.

** & Aljaz, T.

***

* Gorenje Household Appliances, Partizanska 12, 3320 Velenje, Slovenia ** University of Maribor, Faculty of Mechanical Engineering, Smetanova 17, 2000 Maribor, Slovenia

*** Faculty of Information Studies, Ljubljanska 31a, 8000 Novo mesto, Slovenia

E-Mail: [email protected], [email protected]; [email protected]

Abstract

Today, success represents sustainability in introducing new standards and orientation toward the

customer, to the quality and price of products, to flexibility, agility and promptness, to economising in

resources and protection of the environment. This is not easy to achieve, especially in an environment,

where it is natural desire to keep resources busy, in particular the critical resources. This usually

results in unfavourable outcomes, usually experiencing significant degradation in performance of

company and a lot of firefighting inside the company.

This paper outlines three approaches for manufacturing scheduling, starting with traditional push

approach (MRP) and continuing with Kanban (pull) and Theory of Constraints (pull/push), regarding

performance and throughput improvement. It determines how resources should be scheduled within a

system in order to enhance performance, provide stability and predictability of products. Simulations

demonstrate that 200 % to 600 % of improvement is possible at inventory level (work in progress) and

the same at lead time, based on the number of completed tasks over a given period of time. (Received in February 2017, accepted in May 2017. This paper was with the authors 1 month for 1 revision.)

Key Words: Scheduling, Kanban, Theory of Constraints, Comparison, Lead Time, WIP

1. INTRODUCTION

On the market, which is shaped by increasingly more demanding customers, the rising

number of providers and the competitiveness between them, and right business and

production strategy are the deciding factors of a company’s success [1, 2]. Companies are

searching how to efficiently use their resources. Time is a limiting unrecoverable resource and

we must schedule our tasks to utilise this limited resource in an optimum manner. Efficient

use of time is namely one of the greatest indicators of competitiveness. In order to achieve

that, they are heavily loading system's resources, so that the key resources are always busy.

On the other hand, the system is strong as the weakest chain in a system. Therefore, the

capacity or throughput of the system is limited by the weakest link, called capacity

constrained resources (CCR) [3].

Clearly, having less work in the system will result in starvation of CCR and consequently

also in reduced throughput of the system. On the other hand, due dates and quality

requirements of customers are one of the major criteria to retain and build competitive

advantage on markets. However, this is difficult to achieve, if the CCRs are overloaded with

tasks. Therefore, the need for solution is identified that will schedule tasks throughout the

system and will limit the number of active tasks and increase predictability and flow stability

through it.

Traditional, push based approach is usually suggested for products with small demand

uncertainty, as the forecast will provide a good indication of what to produce and keep in

inventory, and also for products with high importance of economies of scale in reducing costs.

The scheduling functions in a company rely on mathematical techniques and heuristic

methods to allocate limited resources to the activities that have to be done. Beside classic

mailto:[email protected]



Zupancic, Buchmeister, Aljaz: Reducing the Time of Task Execution with Existing Resources …

485

dispatching rules to sequence the work waiting at workstations plenty of non-conventional

methods (algorithms using artificial intelligence) are applied for optimization of scheduling

problems [4-9]. Objectives can take many different forms, such as minimizing the time to

complete all activities, minimizing the number of activities that are completed after the

committed due dates, and so on.

Two advanced general approaches have been widely used in the past decades in

manufacturing companies in order to improve throughput of the system, thus reducing

number of active tasks in a system: Kanban and Theory of Constraints (TOC) [10].

Kanban [11], translated signboard or billboard in Japanese, is a scheduling system to

control number of active tasks in a system, developed by Taiichi Ohno, an industrial engineer

at Toyota. Kanban became an effective tool to support running a production system as a

whole, and a way to promote improvement. One key indicator of the success of production

scheduling based on demand, pushing, is the ability of the demand-forecast to create such a

push. Kanban, by contrast, is part of an approach where the "pull" comes from demand. Re-

supply or production is determined according to the actual demand of the customer. In

contexts where supply time is lengthy and demand is difficult to forecast, often, the best one

can do is to respond quickly to observed demand. This situation is exactly what a Kanban

system accomplishes, in that it is used as a demand signal that immediately travels through

the supply chain. Where the supply response is not quick enough to meet actual demand

fluctuations, thereby causing potential lost sales, stock building may be deemed more

appropriate, and is achieved by placing more kanban in the system.

Second approach used in our paper, was introduced by Eliyahu M. Goldratt, is called

Theory of Constraints (TOC) [12]. Despite its name it is not particularly theoretical. Rather,

its tools and applications are designed to solve business problems in a practical and effective

manner. The methodology assumes that every system or organization can be characterized as

network of interdependent processes or elements. This means that systems are analogous to

chains, or networks of chains. Like a chain, the system performance is limited by the weakest

link – constraint. This means that no matter how much effort you put into improving the

processes of a system, only the improvements to the weakest link will produce any detectable

system improvement [13-16]. TOC provides tools and applications that enable organizations

to identify the constraints (or few of them), exploits them and subordinates others based on

that decision in order to get the most out of the existing system or organization. To

accomplish this, TOC shifts the focus of management from optimizing separate organizational

units, functions and resources to increasing the flow of throughput generated by the entire

system. TOC's key processes are focused on removing barriers that prevent each part from

working together as an integrated whole. TOC becomes an important problem structuring and

solving methodology which changes the way of thinking of managers [17-19].

Additionally, the Drum-Buffer-Rope (DBR) [16, 20, 21] is powerful and robust TOC

solution that is intended to manage the flow of work through a (development) process rather

than managing the capacity of resources. It is designed to protect against general cause

variation that cannot be removed from the system and some special cause variation (e.g.,

Murphy). As basis for its work it uses first three steps of five focusing steps defined by TOC:

identify the system constraint, decide how to exploit the system constraint and subordinate

everything else to the above decisions. Basic principle of DBR is shown in Fig. 1.

A lot of researchers have simulated DBR systems, sometimes to estimate DBR’s

parameters, such as the time buffer, or to compare DBR’s effectiveness with other systems.

The analysis presented in this paper offers a simple methodology to help decide when to

release additional work to the system. A discrete-event simulation model was developed to

compare and evaluate traditional manufacturing scheduling approach with Kanban and TOC.

The simulator has been written in C# as a WPF program in Microsoft Visual Studio 2015.


486

Figure 1: Basic principles of Drum-Buffer-Rope (DBR).

1.1 Problem definition

Through the simulation, we want to answer the following questions:

1. What are the average task execution times (lead time) for traditional manufacturing (to

be used as a baseline)?

2. How does the average task execution time (lead time) for Kanban compare to the

average time for traditional manufacturing at the same process?

3. How does the average task execution time for TOC compare to the average time for

traditional manufacturing at the same process?

4. How does the average task execution time for TOC compare to the average time for

Kanban at the same process?

5. Which methodology gives the best results considering average task execution time,

standard deviation for task execution time, and lowest number of active tasks / inventory

(Work-in-Progress – WIP level)?

2. SIMULATION SETUP

As it is described in [22] the Dice Game is intended to demonstrate the influence of the

variability (or statistical fluctuations as they were named by Goldratt [10]). For simulation,

we are using balanced dices (with 4, 6, 8, 12, 16 and 20 surfaces) distributed on workplaces.

Each workplace has one operator and one machine, represented by the player and the dice (in

each cycle player throws the dice ones and the value of the dice represents the production of

that workplace for cycle). Workplaces are ordered in one linear sequence. At the Product

Orders we have unlimited supply of tokens and for Finished Products we assume that they are

all purchased directly from the end of the production line. Each workplace has a WIP storage,

where finished (semi)products from the previous cycle are stored.

Simulation model (for all cases) consists of 6 workplaces (Fig. 2) with balanced dice

assigned as follows:

PLAN – max. 8 products per cycle,

DM1 – max. 20 semi-finished products per cycle,




DM5 – max. 6 finished products per cycle.

Figure 2: Process setup in simulation (during sample simulation in the simulator).


487

Basic rules for the simulation:

Each workplace has on unbalanced dice with different number of surfaces.

One cycle is one dice roll at each workplace representing the productivity for one 8-hour

shift.

In each cycle, all workplaces perform the dice roll at the same time (since they are all

working in the same 8-hour shift).

First workplace (PLAN) plays the role of the gatekeeper and pulls the material from

unlimited supply.

All other workplaces pull needed material from the WIP of previous workplace.

o Each operator can only take the maximum amount available at previous workplace WIP.

o If there is not enough available material in the WIP of previous workplace, the operator

can’t wait for additional material from current simulation step.

We start the simulation with zero material in each workplace WIP.

Other assumptions for the simulation:

We have a simplified process:

o No lead time at the beginning of each cycle,

o At each workplace, we only have one machine (if we have more, they are represented

with higher available maximum productivity and are considered as one, more

productive, machine).

No problems with workforce and logistics:

o No prioritizations of tasks,

o Independent process – no artificial delays,

o All input materials are available on demand,

o No sick leave of workers or mechanical failures on machines,

o Customer buys all finished products immediately.

2.1 Simulation setup for traditional production scheduling

Simulation setup for traditional scheduling is set up with no constrains on WIP with

maximum throughput based on dice roll in each cycle.

Each simulation will run for 100 cycles with 20 replications.

2.2 Simulation setup for kanban production

Simulation setup for kanban production is set up with the constraint of 5 semi-finished

products in WIP for each workplace (estimation based on trial runs).


2.3 Simulation setup for DBR (TOC) production

Based on results of simulation for traditional production scheduling, it is established, that

CCR (Capacity Constrained Resource) is workplace DM4 and it plays the role of the Drum in

simulation of this process. Buffer capacity is set to 15 semi-finished products based on trial

runs.


3. RESULTS AND DISCUSSION

In Table I the following parameters are displayed on the horizontal axis for each process,

based on 20 simulations:

Min. – minimal value among 20 performed simulations,

Max. – maximum value among 20 performed simulations,


488

Avg. – average value among 20 performed simulations.

On vertical axis, the following parameter values are displayed:

End WIP – WIP after the end of simulation,

Finished products – quantity of finished products after 100 cycles of simulation,

Production time, min. – minimum production time in 100 cycles of simulation,

Production time, max. – maximum production time in 100 cycles of simulation,

Production time, avg. – average production time for 100 cycles of simulation,

Standard deviation – standard deviation of production time for 100 cycles of simulation.

Table I: Summary of 60 simulations.

Approach: Traditional Kanban DBR

Min. Max. Avg. Min. Max. Avg. Min. Max. Avg.

End WIP 184 265 217,55 12 17 14 14 18 15,35

Finished products 173 203 186,35 161 181 172,1 172 205 185,25

Production time, min. 5 5 5 5 5 5 5 5 5

Production time, max. 46 62 53,2 9 12 10,45 9 15 11,15

Production time, avg. 25 35 29,1 7 8 7,2 6 8 7,05

Standard deviation 11,77 17,88 13,927 0,76 1,54 1,172 0,91 1,71 1,2285

3.1 Results by used approach

Traditional production scheduling

Simulation of Traditional production scheduling serves as a baseline for comparison with

other approaches, for initial WIP restriction and for CCR identification.

Table II: Traditional production scheduling simulations summary.

Iteration End WIP Finished

Duration

Min. Max. Avg. Std. dev.

1 205 176 5 50 30 13,36

2 240 185 5 54 29 14,13

3 192 195 5 48 27 12,73

4 214 183 5 52 30 14,11

5 248 186 5 60 35 14,19

6 244 173 5 62 31 17,88

7 219 193 5 52 31 13,99

8 195 181 5 55 31 13,54

9 220 189 5 53 27 14,34

10 210 194 5 50 27 13,28

11 184 181 5 47 25 12,64

12 188 186 5 52 29 12,46

13 189 187 5 46 28 11,77

14 199 200 5 50 26 13,26

15 220 186 5 56 32 15,87

16 249 194 5 56 29 14,48

17 265 173 5 56 30 14,13

18 218 180 5 57 29 14,56

19 255 182 5 57 29 14,77

20 197 203 5 51 27 13,05


489

Min. 184 173 5 46 25 11,77

Max. 265 203 5 62 35 17,88

Avg. 217,55 186,35 5 53,2 29,1 13,927

Table II shows that average number of finished products in 100 iterations was 186,35. End

WIP after 100 iterations is 217,55, which represents 177 % of finished products.

Fastest production time was 5 cycles while slowest was on average 53,2 cycles resulting

in production time average of 29,1 cycle with average standard deviation of 13,927 cycles.

Iteration no. 7 is closest to these averages and is used for further more detailed analysis.

Figure 3: Simulation 7 – current cycle WIP at workplaces.

Fig. 3 reveals that WIP is relatively stable around 5 on all workplaces, except before

workplace DM4, where they linearly grow. This means, that our CCR is workplace DM4.

Figure 4: Simulation 7 – production at workplaces.

Fig. 4 displays production (flow) for each workplace by iterations. There are no

production outages (due to lack of material at previous workplace WIP) detected except at the


490

beginning of production (due to lead time). Chart also reveals that DM5 is not working at full

capacity because of the limited supply from the DM4. Production on other workplaces is

within expected statistical variance.

Kanban

Based on the results for Traditional production scheduling, where WIP was around 5 on

almost all workplaces except CCR, this number was used as WIP constraint for each

workplace WIP.

Table III: Kanban simulations summary.

Iteration WIP Finished

Duration


1 13 173 5 11 7 1,53

2 16 172 5 10 7 1

3 14 166 5 10 8 0,99

4 13 176 5 10 7 1,25

5 12 173 5 11 7 1,28

6 14 171 5 11 7 1,08

7 15 177 5 9 7 0,76

8 15 173 5 10 7 0,98

9 12 177 5 9 7 1,03

10 17 165 5 11 7 1,4

11 13 177 5 10 7 1,03

12 14 181 5 10 7 0,9

13 16 171 5 11 7 1,32

14 14 177 5 10 7 1,22

15 13 166 5 10 8 1,14

16 14 172 5 10 7 1,21

17 13 176 5 11 7 1,16

18 12 176 5 11 7 1,54

19 15 161 5 12 8 1,47

20 15 162 5 12 8 1,15

Min. 12 161 5 9 7 0,76

Max. 17 181 5 12 8 1,54

Avg. 14 172,1 5 10,45 7,2 1,172

Table III shows that the average number of finished products in 100 iterations was 172,1

(which represents 92 % of production with Traditional production scheduling). Average end

WIP after 100 iterations is 14, which represents 8 % of finished products. End WIP with

Kanban represents only 3 % of end WIP with Traditional production scheduling.



Iteration no. 16 is closest to these averages and is used for further more detailed analysis.

Fig. 5 reveals that WIP is relatively stable at all workplaces, except at workplace DM5,

where they are the lowest due to limited supply from DM4 which was already identified as

CCR.


production outages (due to lack of material at previous workplace WIP) detected except at the

beginning of production (due to lead time). Chart also reveals that production at all

workplaces is relatively synchronized.


491

Figure 5: Simulation 16 – current cycle WIP at workplaces.


TOC – DBR

Based on results for Traditional production scheduling, where WIP at all workplaces before

CCR was around 5, this value was used for buffer size estimation. Since there are three

workplaces before CCR/DM4 (DM1, DM2 and DM3 – PLAN is in our process only

gatekeeper), size of the buffer was set to 15.

Table IV shows that average number of finished products in 100 iterations was 185,25

(which represents 99 % of production with Traditional production scheduling). Average end

WIP after 100 iterations is 15,35, which represents 8,2 % of finished products.



Iteration no. 18 is closest to this averages and is used for further more detailed analysis.

Fig. 7 reveals that WIP is relatively stable at all workplaces except WIP before DM4

(already identified as CCR) which varies between 3 and 10.


production outages (due to lack of material at previous workplace WIP) detected, except on

the beginning of production (due to lead time). Chart also reveals that production at all

workplaces is synchronized resulting in smooth production flow.


492

Table IV: DBR simulations summary.

Iteration WIP Finished

Duration


1 15 175 5 11 8 1,31

2 16 178 5 11 7 1,44

3 15 194 5 10 7 1,05

4 15 179 5 12 7 1,14

5 15 199 5 11 7 0,99

6 15 172 5 15 8 1,21

7 15 181 5 11 6 1,21

8 17 189 5 10 7 1

9 14 189 5 11 7 1,26

10 15 184 5 12 7 1,28

11 16 183 5 13 7 1,67

12 16 187 5 11 7 1,14

13 16 185 5 11 7 1,37

14 14 205 5 9 6 0,91

15 15 174 5 11 8 1,23

16 18 177 5 12 7 1,71

17 15 188 5 10 7 0,93

18 16 184 5 11 7 1,27

19 14 197 5 10 7 1,15

20 15 185 5 11 7 1,3

Min. 14 172 5 9 6 0,91

Max. 18 205 5 15 8 1,71

Avg. 15,35 185,25 5 11,15 7,05 1,2285

Figure 7: Simulation 18 – WIP before each workplace.


493


3.2 Comparative analysis by used approach

WIP

Since WIP represents the cost in our process and at the same time it acts as a buffer, it is

ideally low, but not too low to cause delays in the manufacturing process.

Fig. 9 displays the area of End WIP values for 20 simulations for each process type. It is

evident that End WIP is highest for Traditional production scheduling and that this approach

has the highest dispersion of results among all three approaches (ranging from 184 to 256).

End WIP for Kanban and DBR is considerably lower and of similar size (Kanban: from 12

to 17; DBR: from 14 to 18) with lower dispersion of results.

Results show that from End WIP size management point Kanban and DBR give better

result.

Figure 9: Quartile chart of End WIP simulation Figure 10: Quartile chart of Finished products

results by used approach. simulation results by used approach.


494

Finished products

Number of finished products provides insight into process throughput and higher values are

more desirable.

Fig. 10 displays the area of values for finished products where we can see a similar

number and dispersion of finished products values for Traditional production scheduling

(from 173 to 203) and DBR (from 172 to 205).

Kanban is slightly lower in this category (from 161 to 181) but has smaller dispersion of

results.

Average production time

Production time is important for product delivery planning, so lower values are more

desirable. Moreover, we also want low standard deviation of production times, so we can plan

product delivery time with higher confidence.

Fig. 11 displays the area of values for average production times. It is clearly noticeable

that these values are highest for Traditional production scheduling (from 25 to 35 cycles) with

highest dispersion. Results for Kanban (from 7 to 8 cycles) and DBR (6 to 8 cycles) are

comparable and considerably lower (slightly higher than theoretical minimum of 5 cycles)

with low dispersion.

Let’s look at the second criteria connected to previous parameter: standard deviation of

production times.

Fig. 12 displays the area of values for standard deviations of production times by process

type. It is evident that predicting production time is hardest for Traditional production

scheduling where standard deviation is from 11,77 to 17,88 cycle with average of 13,927

cycle. Kanban and DBR have comparable results as well. Kanban has standard deviation

ranging from 0,76 to 1,54 with average of 1,172 cycle. DBR has standard deviation from 0,91

to 1,71 cycle, with average of 1,2285 cycle.

Figure 11: Quartile chart of average production Figure 12: Quartile chart of standard deviation

time by used approach. of production times.

With combination of both values the following production times are predicted:

Traditional production scheduling has average production time of 29,1 cycle and average

standard deviation of 13,927 cycle, so we can expect production times from 15 to 43

cycles.


495

Kanban has average production time of 7,2 cycle and average standard deviation of 1,172

cycle, so we can expect production times from 6 to 8 cycles.

DBR has average production time of 7,05 cycles and average standard deviation of 1,2285

cycle, so we can expect production times from 6 to 8 cycles.

4. CONCLUSION

Scheduling as a research area is motivated by questions that arise in production planning and,

generally, in all situations in which scarce resources have to be allocated to activities over

time. Many manufacturing companies are now critically evaluating their processes to

determine their effectiveness in bringing maximum value to customers. They strive to include

methods that greatly minimize lead time, reduce costs, and improve quality.

In this paper it has been analysed and formalized some findings using simulator for

implementation of Kanban and Theory of Constraints tools on sample manufacturing line.

Analyses showed that resource productivity is at its highest with traditional production

scheduling, but on the account of high inventory (number of active tasks) at the Capacity

Constrained Resource (CCR) and by far the longest time for task completion (lead time).

Kanban as well as TOC showed a stable and relatively low number of active tasks and shorter

task completion time compared to traditional production scheduling, but on the account of

lower resource efficiency.

The key finding of this article is that control over the number of active tasks in a system is

more important than continuous full utilization of all resources (efficiency) and significantly

impacts production lead time through lower number of active tasks (effectiveness).

Additionally, it was showed that introduction of control of active tasks is simpler with

Kanban, but TOC offers better results at system fluidity through advanced management of a

buffer, and at the same time it leaves much space for improvement in the future with the

introduction of Fever Chart for prioritization of tasks that are already within a system.

REFERENCES

[1] Bagheri, F.; Noorossana, R. (2016). A statistical model for determination of the type of

knowledge management approach based on organization processes, Transactions of FAMENA,

Vol. 40, No. 3, 43-56, doi:10.21278/TOF.40304

[2] Kapor, N.; Milinovic, M.; Jeremic, O.; Petrovic, D. (2015). Deterministic mathematical

modelling of platform performance degradation in cyclic operation regimes, Strojniski vestnik –

Journal of Mechanical Engineering, Vol. 61, No. 3, 167-175, doi:10.5545/sv-jme.2014.2294

[3] Goldratt, E. M. (2003). Production the TOC Way with Simulator, The North River Press

Publishing Corporation, Great Barrington

[4] Tasič, T.; Buchmeister, B.; Ačko, B. (2007). The development of advanced methods for

scheduling production processes, Strojniski vestnik – Journal of Mechanical Engineering, Vol.

53, No. 12, 844-857

[5] Rathinam, B.; Govindan, K.; Neelakandan, B.; Raghavan, S. S. (2015). Rule based heuristic

approach for minimizing total flow time in permutation flow shop scheduling, Technical Gazette,

Vol. 22, No. 1, 25-32, doi:10.17559/TV-20130704132725

[6] Buchmeister, B.; Friscic, D.; Palcic, I. (2013). Impact of demand changes and supply chain’s

level constraints on bullwhip effect, Advances in Production Engineering & Management, Vol. 8,

No. 4, 199-208, doi:10.14743/apem2013.4.167

[7] Majazi Dalfard, V.; Ranjbar, V. (2012). Multi-projects scheduling with resource constraints &

priority rules by the use of simulated annealing algorithm, Technical Gazette, Vol. 19, No. 3,

493-499

https://doi.org/10.21278/TOF.40304

https://doi.org/10.5545/sv-jme.2014.2294

https://doi.org/10.17559/TV-20130704132725

https://doi.org/10.14743/apem2013.4.167


496

[8] Tang, M.; Gong, D.; Liu, S.; Zhang, H. (2016). Applying multi-phase particle swarm

optimization to solve bulk cargo port scheduling problem, Advances in Production Engineering

& Management, Vol. 11, No. 4, 299-310, doi:10.14743/apem2016.4.228

[9] Li, Y.; Yao, X.; Zhou, J. (2016). Multi-objective optimization of cloud manufacturing service

composition with cloud-entropy enhanced genetic algorithm, Strojniski vestnik – Journal of

Mechanical Engineering, Vol. 62, No. 10, 577-590, doi:10.5545/sv-jme.2016.3545

[10] Goldratt, E. M.; Cox, J. (2004). The Goal: A Process of Ongoing Improvement, North River

Press, Great Barrington

[11] Lu, D. J. (1989). Kanban Just-in Time at Toyota: Management Begins at the Workplace (Volume

1), Productivity Press, Portland

[12] Goldratt, E. M. (1999). Theory of Constraints, North River Press Publishing Corporation, Great

Barrington

[13] Simsit, Z. T.; Gunay, N. S.; Vayvay, O. (2014). Theory of constraints: a literature review,

Proceedings of the 10th International Strategic Management Conference, 930-936

[14] Naor, M.; Bernardes, E. S.; Coman, A. (2013). Theory of constraints: is it a theory and a good

one?, International Journal of Production Research, Vol. 51, No. 2, 542-554,

doi:10.1080/00207543.2011.654137

[15] Golmohammadi, D. (2015). A study of scheduling under the theory of constraints, International

Journal of Production Economics, Vol. 165, 38-50, doi:10.1016/j.ijpe.2015.03.015

[16] Zhang, X. M.; Du, Y. L. (2015). Research of production scheduling based on theory of

constraints, Proceedings of the 2015 International Conference on Electrical, Automation and

Mechanical Engineering, 142-145

[17] Banerjee, A.; Mukhopadhyay, S. K. (2016). A contemporary TOC innovative thinking process in

the backdrop of leagile supply chain, Journal of Enterprise Information Management, Vol. 29,

No. 3, 400-431, doi:10.1108/JEIM-08-2014-0086

[18] Golmohammadi, D.; Mansouri, S. A. (2015). Complexity and workload considerations in product

mix decisions under the theory of constraints, Naval Research Logistics, Vol. 62, No. 5, 357-369,

doi:10.1002/nav.21632

[19] Wang, J.-Q.; Zhang, Z.-T.; Chen, J.; Guo, Y.-Z.; Wang, S.; Sun, S.-D.; Qu, T.; Huang, G. Q.

(2014). The TOC-based algorithm for solving multiple constraint resources: a re-examination,

IEEE Transactions on Engineering Management, Vol. 61, No. 1, 138-146,

doi:10.1109/TEM.2013.2264830

[20] Aljaž, T. (2014). Out of chaos in 12 months – improving lead time of sprint projects in software

development implementing Drum Buffer Rope Solution, ERK'2014 Conference Proceedings, 48-

51

[21] Chakravorty, S. S.; Hales, D. N. (2016). Improving labour relations performance using a

Simplified Drum Buffer Rope (S-DBR) technique, Production Planning & Control, Vol. 27, No.

2, 102-113, doi:10.1080/09537287.2015.1079744

[22] Schragenheim, E.; Dettmer, H. W. (2001). Manufacturing at Warp Speed: Optimizing Supply

Chain Financial Performance, CRC Press, Boca Raton

https://doi.org/10.14743/apem2016.4.228

https://doi.org/10.5545/sv-jme.2016.3545

https://doi.org/10.1080/00207543.2011.654137

https://doi.org/10.1016/j.ijpe.2015.03.015

https://doi.org/10.1108/JEIM-08-2014-0086

https://doi.org/10.1002/nav.21632

https://doi.org/10.1109/TEM.2013.2264830

https://doi.org/10.1080/09537287.2015.1079744

Documents

REDUCING THE TIME OF TASK EXECUTION WITH EXISTING